{"id":"https://openalex.org/W7223355807","doi":"https://doi.org/10.48550/arxiv.2610.31264","title":"Deep ReLU Networks Can Fail the Best Approximation Property for Continuous Target Functions","display_name":"Deep ReLU Networks Can Fail the Best Approximation Property for Continuous Target Functions","publication_year":2026,"publication_date":"2026-10-22","ids":{"openalex":"https://openalex.org/W7223355807","doi":"https://doi.org/10.48550/arxiv.2610.31264"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.31264","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.31264","pdf_url":"https://arxiv.org/pdf/2610.31264","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.31264","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5045569270","display_name":"Sebastian Kassing","orcid":"https://orcid.org/0000-0002-2016-899X","observed_orcids":["https://orcid.org/0000-0002-2016-899X"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Kassing, Sebastian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5045569270"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9848999977111816,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9848999977111816,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.00930000003427267,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12676","display_name":"Machine Learning and ELM","score":0.0012000000569969416,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/relu-neural-networks","display_name":"ReLU networks","score":0.8859999775886536},{"id":"https://openalex.org/keywords/counterexample","display_name":"counterexample","score":0.859000027179718},{"id":"https://openalex.org/keywords/deep-neural-network","display_name":"deep neural network","score":0.7039999961853027},{"id":"https://openalex.org/keywords/breaklines","display_name":"breaklines","score":0.671999990940094},{"id":"https://openalex.org/keywords/global-minimum","display_name":"global minimum","score":0.6460000276565552},{"id":"https://openalex.org/keywords/shallow-neural-networks","display_name":"shallow neural networks","score":0.6389999985694885},{"id":"https://openalex.org/keywords/symmetry-arguments","display_name":"symmetry arguments","score":0.3709999918937683}],"concepts":[{"id":"https://openalex.org/C162838799","wikidata":"https://www.wikidata.org/wiki/Q596077","display_name":"Counterexample","level":2,"score":0.8198999762535095},{"id":"https://openalex.org/C95611797","wikidata":"https://www.wikidata.org/wiki/Q17502105","display_name":"Infimum and supremum","level":2,"score":0.7825000286102295},{"id":"https://openalex.org/C186633575","wikidata":"https://www.wikidata.org/wiki/Q845060","display_name":"Maxima and minima","level":2,"score":0.7495999932289124},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.6513000130653381},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.569100022315979},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5654000043869019},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.4765999913215637},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.4542999863624573},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41839998960494995},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.4074999988079071},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4009000062942505},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.37209999561309814},{"id":"https://openalex.org/C152306953","wikidata":"https://www.wikidata.org/wiki/Q621744","display_name":"Approximation property","level":3,"score":0.3653999865055084},{"id":"https://openalex.org/C2779886137","wikidata":"https://www.wikidata.org/wiki/Q21030012","display_name":"Symmetry (geometry)","level":2,"score":0.36320000886917114},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.3490999937057495},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.33059999346733093},{"id":"https://openalex.org/C2778199601","wikidata":"https://www.wikidata.org/wiki/Q5165476","display_name":"Continuous function (set theory)","level":3,"score":0.3140999972820282},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29809999465942383},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.29249998927116394},{"id":"https://openalex.org/C27142425","wikidata":"https://www.wikidata.org/wiki/Q1139334","display_name":"Almost everywhere","level":2,"score":0.2906000018119812},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.27889999747276306},{"id":"https://openalex.org/C145242015","wikidata":"https://www.wikidata.org/wiki/Q774123","display_name":"Approximation theory","level":2,"score":0.27549999952316284},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.2660999894142151}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.31264","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.31264","pdf_url":"https://arxiv.org/pdf/2610.31264","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.31264","doi":"https://doi.org/10.48550/arxiv.2610.31264","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.31264","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.31264","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.31264","pdf_url":"https://arxiv.org/pdf/2610.31264","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7223355807.pdf","grobid_xml":"https://content.openalex.org/works/W7223355807.grobid-xml"},"referenced_works_count":2,"referenced_works":["https://openalex.org/W4310430694","https://openalex.org/W7161718229"],"related_works":["https://openalex.org/W4293084530","https://openalex.org/W4392788178","https://openalex.org/W2766099245","https://openalex.org/W3087523251","https://openalex.org/W2787302432","https://openalex.org/W2755888403","https://openalex.org/W3198891595","https://openalex.org/W3135609499","https://openalex.org/W2951316532","https://openalex.org/W4394671854"],"abstract_inverted_index":{"We":[0],"construct":[1],"a":[2],"continuous":[3,49],"function":[4],"on":[5,86],"$[-1,1]^2$":[6],"for":[7,48,58,92],"which":[8,101],"the":[9,12,16,41,67,73,87,90,95,103,108],"infimum":[10],"of":[11,18,66,75],"$L^2$-approximation":[13],"error":[14],"over":[15],"class":[17],"ReLU":[19,36,63],"networks":[20,37],"with":[21,55],"two":[22],"hidden":[23,97],"layers":[24],"and":[25],"architecture":[26],"$(2,2,1,1)$":[27],"is":[28,64,84],"not":[29,39],"attained.":[30],"This":[31],"counterexample":[32],"shows":[33],"that":[34,71,89],"deep":[35],"do":[38],"satisfy":[40],"best":[42],"approximation":[43],"property":[44],"in":[45,78,94,107],"general,":[46],"even":[47],"target":[50],"functions.":[51],"Our":[52],"findings":[53],"contrast":[54],"existing":[56],"results":[57],"shallow":[59],"neural":[60],"networks,":[61],"where":[62],"one":[65],"few":[68],"activation":[69],"functions":[70],"guarantee":[72],"existence":[74],"global":[76],"minima":[77],"typical":[79],"regression":[80],"tasks.":[81],"The":[82],"proof":[83],"based":[85],"fact":[88],"breaklines":[91],"neurons":[93],"second":[96],"layer":[98],"can":[99],"bend,":[100],"breaks":[102],"symmetry":[104],"arguments":[105],"used":[106],"one-hidden-layer":[109],"case.":[110]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-24T00:00:00"}